The Cylinder Escapement is a frictional-rest watch escapement in which a hollow steel or ruby cylinder mounted on the balance staff replaces the verge's pallets, with the escape-wheel teeth riding directly on the cylinder's inner and outer surfaces. Where the older verge escapement needed a tall vertical crown wheel, the cylinder lies flat and lets the movement sit thin. Each tooth gives a brief impulse as it enters and leaves the cylinder, sustaining the balance. George Graham perfected it in 1726, and it powered millions of slim 19th-century French pocket watches.
Cylinder Escapement Interactive Calculator
Vary cylinder geometry and beat rate to see bore size, lip opening clearance, beat frequency, and escape-wheel speed.
Equation Used
The cylinder bore is estimated from outside diameter minus twice the wall thickness. The opening margin shows how far the cut-away exceeds 180 degrees, split equally between the two lips. Escape-wheel speed assumes a full tooth pitch is used every two beats.
- Cylinder wall is uniform around the remaining arc.
- Opening angle is measured around the cylinder cut-away.
- One escape-wheel tooth pitch is completed every two beats.
- Friction and wear are not included in the numeric outputs.
How the Cylinder Escapement Works
The Cylinder Escapement, also called the horizontal escapement in older English texts, works by routing the escape-wheel teeth straight through a slotted hollow cylinder that forms part of the balance staff itself. The escape wheel sits below the balance, its 15 wedge-shaped teeth raised on stalks so each tooth tip rides at the height of the cylinder's mouth. As the balance swings, a tooth drops onto the outer lip of the cylinder, slides along the curved face giving impulse, then falls inside the cylinder where it rests against the inner wall until the balance returns and lets it escape out the other side. That sliding contact during rest is what defines a frictional rest escapement — the wheel never fully unlocks, it just shifts which surface it's pressing on.
Geometry has to be exact. The cylinder bore must match the tooth-tip radius within roughly 0.01 mm, the cylinder wall thickness sits around 0.10-0.12 mm on a typical 10-ligne movement, and the cut-away opening covers slightly more than 180° of the cylinder's circumference so the tooth can enter and exit on the same swing. If the wall is too thick the tooth won't drop cleanly and the watch sets — stops dead at a specific position every cycle. If the bore is oversized the tooth rattles and you lose amplitude. If the cylinder is steel rather than ruby and the oil dries, the teeth cut grooves into the inner wall within a year or two and rate becomes erratic.
The trade is simple. You get a flat movement and an inherently safe escapement that won't overbank, but you pay for it in friction losses — the escape wheel is in constant rubbing contact with the cylinder, so amplitude rarely exceeds 220° and isochronism is poor compared with a lever. That's why no serious chronometer ever used one, and why almost every cylinder watch you'll meet today is either a slim French dress piece or a budget Swiss product from before 1900.
Key Components
- Cylinder (on balance staff): A hollow tube, typically 0.6-1.2 mm outside diameter, cut away over slightly more than half its circumference to form the entry and exit lips. Wall thickness runs 0.10-0.12 mm in steel or 0.15-0.20 mm in ruby. The cylinder IS the balance staff in this design — it rotates with the balance and receives impulse directly.
- Escape wheel: Usually 15 teeth, each tooth raised on a thin stalk so the tip clears the cylinder body and engages only the lips. Tooth tips are wedge-profiled with the impulse face angled around 30° to the radial. Pitch diameter is matched to the cylinder bore so drop is around 1° of escape-wheel rotation per beat.
- Balance wheel and hairspring: Provides the oscillating mass and restoring torque. Beat rate is typically 18,000 bph for English work and up to 21,600 bph for late Swiss cylinder calibres. Amplitude is limited to roughly 180-220° because of the constant frictional rest — push it higher and the tooth re-enters before the balance has finished its swing.
- Banking pins or banking shoulders: Limit balance amplitude to prevent the cylinder lip from striking a tooth on its back face. Set so that maximum swing leaves at least 5° of safety clearance between the lip and the next tooth at full amplitude.
- Jewelling (when fitted): Better-grade cylinder watches use a ruby cylinder rather than hardened steel, eliminating the wear grooves that plague steel cylinders after 5-10 years. Endstones cap the balance pivots; the escape-wheel pivots run in plain bearings on cheap calibres and jewelled holes on better ones.
Who Uses the Cylinder Escapement
The Cylinder Escapement (perspective) earned its place wherever thinness mattered more than rate accuracy. From the 1720s onward it became the default escapement for slim dress pocket watches, and through the 19th century the Swiss and French industrial workshops produced it by the million. You almost never see it specified for new work today, but as a restorer or collector you'll meet it constantly — and getting the cylinder bore, escape-wheel tooth height, and balance amplitude all matched is the whole job.
- Horology — antique restoration: Lépine calibre French pocket watches from roughly 1790-1880, where the flat layout pioneered by Jean-Antoine Lépine depended on the cylinder escapement to keep the movement under 4 mm thick.
- Horology — English work: Graham-pattern English pocket watches from 1726 onward, where George Graham's own workshop refined the cylinder escapement (escape-wheel detail) into a form that ran reliably enough to dominate London output for two generations.
- Horology — Swiss industrial: Roskopf-era and pre-Roskopf budget Swiss watches from the 1860s-1900s, including huge volumes of cylindre stamped on the movement bridge by makers in the Vallée de Joux and around La Chaux-de-Fonds.
- Museum and teaching collections: Working demonstration models at institutions like the British Museum and the Musée international d'horlogerie in La Chaux-de-Fonds, where the cylinder's visible action makes it useful for showing frictional rest behaviour to students.
- Collector market: Pre-1900 ladies' pendant watches and chatelaine watches, almost universally cylinder-escapement because the slim caliper depth — often under 3.5 mm — could not accommodate a lever and detached escape wheel.
- Replica and homage horology: Modern small-series replicas of 18th-century watches built by independent restorers, where period correctness demands a cylinder escapement rather than a substituted lever.
The Formula Behind the Cylinder Escapement
What a restorer actually wants to compute is the beat rate the movement will hold given the balance, hairspring, and escape-wheel tooth count. The relationship is fixed by the escapement: each tooth produces two beats (one on entry, one on exit), so the beats per hour follow directly from balance frequency. At the low end of typical cylinder work — 14,400 bph on early English Graham pieces — the watch ticks slowly enough that you can count beats by ear, but rate sensitivity to position is high. At the nominal 18,000 bph of mainstream 19th-century French and English cylinder watches, you hit the sweet spot for cylinder geometry: enough impulse frequency to average out positional error, low enough amplitude demand that the frictional rest doesn't crush isochronism. Push to 21,600 bph in late Swiss calibres and you're fighting the escapement — amplitude drops, oil shears faster, and the rate becomes more sensitive to mainspring state of wind.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| bph | Beats per hour of the movement | beats/hour | beats/hour |
| Nteeth | Number of teeth on the escape wheel (typically 15 for cylinder work) | teeth | teeth |
| fbalance | Balance oscillation frequency (full cycles per second) | Hz | cycles/s |
| Tbalance | Period of balance oscillation, equal to 1 / fbalance | s | s |
Worked Example: Cylinder Escapement in an 1845 Lépine-calibre French pocket watch
A private collector hands you an 1845 Lépine-calibre French pocket watch signed Bréguet à Paris with a 15-tooth steel escape wheel and a damaged hairspring. You need to specify the replacement hairspring stiffness so the watch beats at the original rate. You measure the balance moment of inertia at 8.5 × 10⁻¹⁰ kg·m² and confirm the escape wheel has 15 teeth. The customer wants the watch running at the period-correct 18,000 bph.
Given
- Nteeth = 15 teeth
- Target bph = 18000 beats/hour
- Ibalance = 8.5 × 10⁻¹⁰ kg·m²
Solution
Step 1 — at the nominal target of 18,000 bph, solve the beat equation for balance frequency:
The factor of 2 comes from two beats per balance cycle (one entry impulse, one exit impulse). The escape-wheel tooth count sets the gear train ratios upstream but does not enter the beat equation directly — the balance itself defines beat rate.
Step 2 — at nominal, compute the required hairspring stiffness from the balance period:
Step 3 — at the low end of the cylinder operating range (14,400 bph, typical of early Graham work), the same balance would need:
That is a noticeably softer hairspring — the watch ticks at a slow, audible 4 beats per second and you can almost count them. At the high end, 21,600 bph from a late Swiss cylinder calibre, the same balance demands khigh = 3.02 × 10⁻⁷ N·m/rad, a 44% stiffer spring. In practice that pushes amplitude down by 15-20° because the cylinder's frictional rest cannot extract enough energy per beat to maintain the higher swing rate.
Result
The replacement hairspring needs a torsional stiffness of 2. 10 × 10⁻⁷ N·m/rad to deliver the period-correct 18,000 bph. In practice that means selecting a Nivarox-1 grade alloy spring of around 0.030 mm thickness and 12-13 active turns for a balance of this size, which will give an amplitude in the 200-220° range — exactly what the cylinder geometry expects. Across the range, 14,400 bph runs slow and lazy with high positional error, 18,000 bph is the sweet spot for this calibre, and 21,600 bph chokes amplitude and shortens service life. If after fitting you measure a rate that runs fast by 30+ s/day at full wind but correct at half wind, suspect a hairspring that is touching the regulator pins or the underside of the balance cock — not stiffness error. If the rate drifts by more than 60 s/day between dial-up and pendant-down positions, the cylinder bore is likely oversized relative to the tooth tips, letting the wheel rattle on impulse. And if amplitude is below 180° at full wind, the most common cause is dried oil on the cylinder lips gumming the frictional rest, not a weak mainspring.
Choosing the Cylinder Escapement: Pros and Cons
The Cylinder Escapement sits between the verge it replaced and the lever escapement that eventually replaced it. Each one solves a different problem, and the choice for any restoration or new build comes down to whether thinness, accuracy, or simplicity matters most.
| Property | Cylinder Escapement | Verge Escapement | Lever Escapement |
|---|---|---|---|
| Typical beat rate | 14,400-21,600 bph | 12,000-16,000 bph | 18,000-36,000 bph |
| Daily rate accuracy (well adjusted) | ±30-60 s/day | ±60-120 s/day | ±5-15 s/day |
| Movement thickness achievable | 3-5 mm (slim dress watch) | 8-12 mm (full-plate) | 4-7 mm typical |
| Amplitude range | 180-220° | 90-110° | 270-310° |
| Friction class | Frictional rest (constant contact) | Frictional rest (constant contact) | Detached (free during balance swing) |
| Service life before tooth wear shows | 5-10 years (steel cylinder), 30+ years (ruby) | 10-20 years | 20-50 years |
| Era of dominant use | 1726-1900 | 1500-1750 | 1850-present |
| Restoration complexity | High — cylinder hard to source | Very high — most parts hand-made | Moderate — parts widely available |
Frequently Asked Questions About Cylinder Escapement
Yes — they are the same mechanism. English writers from the 18th and 19th centuries called it the horizontal escapement because the escape wheel lies flat under the balance, contrasting with the vertical crown wheel of the verge it replaced. By the late 19th century cylinder escapement became the dominant term, especially in Continental work. If you are reading a Graham-era treatise and see horizontal escapement, that is your cylinder.
The frictional rest is your culprit. A cylinder escapement consumes energy continuously through the rub of tooth on cylinder wall, while a lever escapement is detached during the balance swing. As mainspring torque drops over the day, the cylinder reaches a threshold where impulse can no longer overcome the rest friction and the watch sets.
Check mainspring tension first — many replacement mainsprings are wound to lever-watch torque levels, which is too weak for cylinder work. The original spring on a cylinder pocket watch was deliberately stronger to compensate for the higher running friction.
Bore-to-tooth-tip mismatch. The cylinder bore must fit the escape-wheel tooth tip radius within about 0.01 mm. If the bore is even 0.02-0.03 mm oversize the tooth has lateral play during impulse, and you hear it as a high-frequency chatter on top of the normal tick.
Measure the tooth-tip pitch radius across at least three teeth (escape wheels often have one or two bent teeth that throw off a single measurement) and compare with the cylinder inside diameter. If the bore is too large, the cylinder is wrong for that wheel — there is no shimming fix that survives more than a few weeks of running.
Ruby, every time, if the budget allows. A steel cylinder running on dried oil cuts visible wear grooves into its inner wall within 12-24 months of daily wear, and once those grooves form the rate becomes erratic in a way no service can fix without replacing the cylinder again. Ruby is essentially immune to that wear mode.
The exception is a museum-grade restoration where period-correct materials matter more than long-term wearability — in that case, match what was originally fitted and tell the customer the watch is a display piece.
Cylinder escapements have inherently poor positional rate because the friction at the cylinder lips changes with the direction gravity loads the balance pivots. In pendant-up the pivot loading shifts the friction couple in a way that drags the balance asymmetrically, and the rest-friction dominance amplifies that small asymmetry into a large rate change.
You cannot adjust this out the way you would on a lever watch — the escapement architecture itself sets the floor for positional error at roughly ±60 s/day. If the customer expects lever-grade accuracy, the honest answer is that the watch cannot deliver it without re-escaping the movement, which destroys its value.
Only if you replace the cylinder, the balance, and recalculate the train. Tooth count interacts with cylinder bore (because tooth-tip pitch radius changes), with balance frequency choice, and with the gear-train ratios upstream that deliver the right wheel speed. Swapping tooth count without redesigning the surrounding parts gives you a watch that either won't run or runs at a wildly wrong rate.
For a 15-tooth original, source a 15-tooth replacement — period French and Swiss cylinder wheels turn up on the parts market regularly, and a competent restorer can re-pivot a worn one rather than substitute a wrong tooth count.
References & Further Reading
- Wikipedia contributors. Cylinder escapement. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
